Full text loading...
- ISSN: 1354-0793
- E-ISSN:
-
Sea Lion Field, North Falkland Basin: seismic inversion and quantitative interpretation
- Source: Petroleum Geoscience, Volume 21, Issue 2-3, Jul 2015, p. 151 - 169
-
- 10 Jul 2014
- 27 Jan 2015
- 27 Mar 2015
Abstract
Extensive use has been made of seismic-derived quantitative interpretation for mapping and predicting reservoir properties in the Sea Lion Field, North Falkland Basin. The Sea Lion sands are acoustically hard compared to the encasing shales, with fairly constant rock properties. The fans are typically 20–60 m in thickness and are predominantly seismically tuned events. Detection of hydrocarbon fluid effects from seismic data has proved elusive to date.
Extended elastic impedance (EEI) was selected as the appropriate seismic analysis tool. Almost all sands are visible on the full-stack seismic data. Coloured inversion of an EEI volume with a rotation angle (χ) of −70° (henceforth shown as EEI(−70)) was found to be an optimal sand predictor and is primarily sensitive to the shale volume of the sands. This EEI χ angle also allows detection of sand that is not observed on the full-stack seismic.
The EEI(−70) coloured inversion data were also used to estimate net sand thickness and the net-to-gross maps of the fans. Subsequently, a new algorithm termed DT-AMP was used to apply seismic amplitude detuning to the full three-dimensional (3D) seismic volume. This has proved useful in the evaluation of new exploration targets beyond the Sea Lion Field area.
A sand classification was performed using multiple EEI attributes, including lithology and fluid cubes. This indicates that there are differences in the amplitude v. offset (AVO) response across the main Sea Lion fan complex.
Pre-stack simultaneous inversion was also performed using both deterministic and stochastic schemes. A three-term inversion was tested but density estimation was considered to be unreliable.
Seismic data were tested at several stages to constrain the static reservoir model, using coloured, deterministic and stochastic inversion methods. The final static model incorporated a set of scenarios using net-to-gross maps inferred from the quantitative analysis of the EEI(−70) data.